The present invention relates to a high speed digital telemetry system for use in an implantable device, such as an implantable pacemaker. More particularly, the present invention relates to an implantable high bit rate telemetry transmitter and corresponding external receiver that transmits data efficiently at a lower than Nyquist bandwidth.
Implantable devices are implanted in a human or animal for the purpose of performing a desired function This function may be purely observational or experimental in nature, such as monitoring certain body functions; or it may be therapeutic or regulatory in nature, such as providing critical electrical stimulation pulses to certain body tissue, nerves or organs for the purpose of bringing about a desired response. The modern implantable demand pacemaker is an example of an implantable device that performs both observational and regulatory functions, i.e., it monitors the heart to make sure it beats at appropriate times; and if not, it causes an electrical stimulation pulse to be delivered to the heart in an attempt to force the heart to beat at an appropriate rate.
In order for an implantable device to perform its functions at minimum inconvenience and risk to the person or animal within whom it is used, some sort of noninvasive telemetry means must be provided that allows data and commands to be easily passed back and forth between the implantable device and an external device. Such an external device, known by a variety of names, such as a controller, programmer, or monitor, provides a convenient mechanism through which the operation of the implantable device can be controlled and monitored, and through which data sensed or detected by the implantable device can be transferred out of the implantable device to an external (non-implanted) location where it can be read, interpreted, or otherwise used in a constructive manner.
As the sophistication and complexity of implantable devices has increased in recent years, the amount of data that must be transferred between an implantable device and its accompanying external device has dramatically increased. This, in turn, has resulted in a search for more efficient ways of effectuating such a data transfer. Any such system must, of course, not only transfer the desired data without significant error, but it must do so rapidly while preserving the limited power resources (e.g., a battery having a limited operating life) of the implanted device.
Heretofore, three basic techniques have been used for communicating with an implantable device: (1) static magnetic field coupling; (2) reflected impedance coupling; and (3) RF coupling. In static magnetic field coupling, a static magnetic field is generated external to the implanted device, e.g., using a permanent magnet, having sufficient strength to close (or open) a magnetic reed switch within the implanted device. While such a technique provides a fairly reliable mechanism for turning various functions within the implanted device ON or OFF, such as turning the telemetry circuits within an implanted device ON only when an external telemetry head is positioned a few inches from the implanted device, the technique is much too slow for efficiently transferring any significant amount of data. Further, for all practical purposes, the static magnetic system is useful only for transferring commands or data to the implanted device, not for transferring data or commands from the implanted device. This is because the weight and/or power requirements associated with the types of permanent magnets or electromagnets needed to operate a magnetic reed switch several inches distant therefrom is incompatible with the requirements of most implantable devices.
In a reflected impedance coupling system, information is transferred using the reflected impedance of an internal (implanted) L-R or L-C circuit energized by an inductively coupled, external L-R or L-C circuit. Such a system is shown, for example, in U.S. Pat. No. 4,223,679. Advantageously, such a system uses little or no current to transmit information. Disadvantageously, the speed at which the information is transferred is quite limited. The external circuit uses an RF (radio frequency) magnetic field carrier. In the cited patent, a voltage controlled oscillator (VCO), in the implanted device, is controlled by the signal to be telemetered. The VCO, in turn, varies the impedance that is reflected. If the signal controlling the VCO is a binary digital signal (having two possible values, e.g., a binary "1" and a binary "0"), this signal encodes the VCO so that the VCO varies from one frequency (representing a binary "1") to another frequency (representing a binary "0"). This technique is known as frequency shift keying (FSK). Each bit duration, i.e., the time in which the binary digit (bit) is expressed, requires a number of carrier cycles. Hence, the bit rate cannot generally be much higher than 10% to 30% of the VCO center frequency. On the other hand, the RF carrier frequency cannot be too high because of the metal enclosure of the implanted device acts as a low pass, single pole filter having an upper cut-off frequency of between 10-30 kHz. Further, the external oscillator L-C circuit typically has a Q (quality factor) of 20 to 50, meaning that the useful modulation bandwidth is limited to around 2 to 5 percent of the RF carrier frequency. This means that a 36 kHz carrier is typically only able to transmit data at a data rate of from 72 to 540 bits per second (bps). Such a rate is generally considered inadequate for modern implantable devices, which devices may have thousands or tens of thousands of bits of data to be transmitted.
In an RF coupled system, information is transferred from a transmitting coil to a receiving coil by way of a carrier signal. The carrier signal is modulated with the data that is to be transmitted using an appropriate modulation scheme, such as FSK or PSK (phase-shift keying--reversing the phase of the carrier by 180 degrees). The modulated carrier induces a voltage at the receiving coil that tracks the modulated carrier signal. This received signal is then demodulated in order to recover the transmitted data. Because of the metal enclosure of the implanted device, which acts as a low pass filter (attenuating high frequencies), the carrier frequency cannot be increased above approximately 10-20 kHz without an unacceptable increase in transmitting coil power. Further, depending upon the type of modulation/demodulation scheme employed, the data or bit rate cannot exceed a prescribed fraction of the carrier frequency, without exceeding a specified amount of mutual interference, i.e., without being able to reliably distinguish between a modulation that represents a binary "1" and modulation that represents a binary "0".
The maximum data transfer rate (bit rate) at which independent signal values can be transmitted over a specified channel without exceeding a specified amount of mutual interference is referred to as the "Nyquist rate." The maximum allowable Nyquist rate is directly related to the bandwidth of the channel through which the data is transferred. Conversely, the "Nyquist bandwidth" is that bandwidth required to allow independent signal values to be transmitted at a given rate without exceeding the specified levels of mutual interference. For example, if the bandwidth of the channel through which the data is transferred is W, the Nyquist rate (assuming an ideal channel) may be as high as 2W. Stated differently, if the data rate is 2W, the Nyquist bandwidth must be at least W. Because of these and other limitations, prior art implantable devices using RF coupling have generally not been able to transfer data at rates in excess of 2-4 kbps. (Note, that as used here and throughout this application, a one-sided bandwidth definition is used. That is, a bandwidth W refers to a range of frequencies from 0 to W, or from -W to 0. Where a carrier signal having a frequency f.sub.c is used, the one-sided bandwidth W refers to a range of frequencies from f.sub.c to f.sub.c +W, or from f.sub.c -W to f.sub.c.)
A further problem affecting the rate at which data can be transferred from an implantable device is electrical noise and/or EMI (electromagnetic interference). In particular, there are at least two primary sources of EMI associated with commonly used types of external devices that significantly affect the range of carrier frequencies and data rates that can be reliably and efficiently (at low power consumption levels) used to transfer data in an RF-type system. First, the input power line frequency (50-60 Hz) of the external device, and the associated switching magnetic fields (e.g., 30 Hz) used with a cathode ray tube (CRT) display, frequently used with external devices, create sufficiently large EMI harmonics to be troublesome as high as 2-6 kHz. Similarly, the 16 kHz line frequency of the horizontal scan of the cathode ray tube (CRT) commonly used with many electronic terminals, makes it extremely difficult to efficiently use a carrier frequency of 16 kHz or higher. In order to minimize the effect of such EMI on the transmission of data from an implanted device used in an environment where such interference is prevalent, and in order to maximize the speed at which the large amounts of data used with modern implantable devices may be transferred, it would thus be preferable to employ a narrow band telemetry channel (to filter out as much EMI and noise as possible) using a carrier signal in the 6-12 kHz range, and using a modulation scheme that permits a data bit rate as high as possible through such channel. While a telemetry system that increases the data bit rate of an implantable device up to an acceptably fast level has heretofore been proposed by the applicant herein, see U.S. Pat. application Ser. No. 07/085,914, filed 08/14/87, such was accomplished only at the expense of a wider bandwidth, thereby causing an increased susceptibility to EMI. What is needed, therefore, is an improved telemetry system for use with an implanted device that not only allows data to be transferred at an acceptably fast rate, e.g., 8 kHz, but that also allows the data at this fast rate to be transferred through a narrow bandwidth, thereby decreasing the susceptibility of the system to EMI and other noise sources. The present invention advantageously addresses this and other needs.
Further, the telemetry system described in the above-cited patent application utilizes an encoding scheme that requires the instantaneous frequency of a carrier signal applied to a transmitting coil to be suddenly changed or reversed. This is accomplished only at the expense of increased complexity in the modulation circuits and drive circuits that apply the modulated carrier signal to the transmitting coil. Such circuits, for example, employ two multiple capacitors that are switchably connected to the transmitting coil. A capacitor connected to the transmitting coil is needed to tune the transmitting coil to the correct frequency. Two capacitors are needed because the voltage applied to the coil by the drive circuits must regularly reverse its polarity, and it takes too much time, and is thus inefficient, for one capacitor to discharge one voltage polarity and charge to another each time that a reversal in the voltage applied to the coil occurs.
Unfortunately, as such telemetry circuits become more and more complex, more and more components are required for the circuits to perform their intended function. As the number of components increases, the cost of the circuit increases and the reliability of the circuit decreases. Therefore, in order to reduce the cost and improve the reliability, simpler and less complex circuits are needed. This is true not only in the implantable transmitter portion of the telemetry system, but also in the external receiver portion of the system where heretofore complex phase locked loops (PLL's) and similar complex circuits have been required in order to demodulate the recovered telemetry signal.